Statement of the Problem The ability to measure local alveolar pAO2 has long been one of the holy grails of pulmonary physiology. Any pulmonary disease necessarily affects either local ventilation or perfusion, or both, and since the local pAO2 is the product of a complicated interplay of the local ventilation and perfusion, it will reflct the mismatch between the two, thus local pAO2 is certainly clinically important. Currently the diagnostic pulmonary V/Q scans are part of the nuclear medicine scans using ionizing radiation. Specific Aims In this application we propose to develop a new non-invasive approach to measure the local alveolar partial pressure of oxygen, pAO2, using hyperpolarized xenon and to evaluate its sensitivity, reproducibility and clinical relevance in the following steps: (SA1) testour hypothesis that the measured pAO2 will be in better agreement with the physiological values of pAO2 when the xenon signal loss due to the gas exchange is measured and accounted; assess its clinical utility through determination of the sensitivity and reproducibility of the technique n healthy volunteers and patients with pulmonary embolism (PE); (SA2) correlate the measured pAO2 in PE patients with V/Q scans obtained from nuclear medicine as part of their clinical workup; compare the ventilation weighted mean value of the measured pAO2 with the global pAO2 value measured using gas equations; test our hypothesis that mean pAO2 values in PE patients will be elevated compared to healthy volunteers, while its spatial distribution in the patients with distal PE will be more heterogeneous compared to the distribution in the patients with proximal PE. Experimental Approach T1 dependence of hyperpolarized Xe on oxygen concentration has been used to obtain pAO2, however no attempt has been made to date to account for the loss of signal due to the gas exchange during the measurement. In this application we will obtain and correct for the gas exchange effect while simultaneously measuring pAO2. The technique will be calibrated in phantoms and in vivo through pAO2 measurements for different O2 concentrations. To assess the reproducibility we will measure pAO2 in 10 volunteers and 10 patients each with proximal and distal PE at least 3 times; normal pAO2 range will be established based on the volunteers data. To assess clinical utility we will correlate patients' pAO2 maps with their V/Q scans, compare mean pAO2 with the global pAO2 obtained using a breath-by breath metabolic cart with directly measured RER and a simultaneously sampled PaCO2, and compare the distribution width of the pAO2 values in patients with proximal and distal PE - we believe it should be wider in the latter group. Significance of the results Knowledge of regional pAO2 is instrumental for the evaluation of V/Q mismatch in pulmonary disease. Although it has been almost a decade since the demonstration of HPXe measurement of pAO2, this is the 1st attempt to account for the gas exchange effect on the measurement. Further, it is still unclear whether this technique is sensitive and reproducible enough to have clinical utility, which we will explore here.